Merge pull request #1414 from ShiningZhang/dev-bug

fix: use_calib is not configured properly
PaddlePaddle · Oct 11, 2021 · 086605b · 086605b
2 parents 32ada7a + 072114e
commit 086605b
Show file tree

Hide file tree

Showing 3 changed files with 3 additions and 3 deletions.
diff --git a/README.md b/README.md
@@ -188,7 +188,7 @@ python3 -m paddle_serving_server.serve --model uci_housing_model --thread 10 --p
 | `use_lite` (Only for Intel x86 CPU or ARM CPU) | -    | -       | Run PaddleLite inference                              |
 | `use_xpu`                                      | -    | -       | Run PaddleLite inference with Baidu Kunlun XPU        |
 | `precision`                                    | str  | FP32    | Precision Mode, support FP32, FP16, INT8              |
-| `use_calib`                                    | bool | False   | Only for deployment with TensorRT                     |
+| `use_calib`                                    | bool | False   | Use TRT int8 calibration                              |
 | `gpu_multi_stream`                             | bool | False   | EnableGpuMultiStream to get larger QPS                |
 
 #### Description of asynchronous model

diff --git a/README_CN.md b/README_CN.md
@@ -187,7 +187,7 @@ python3 -m paddle_serving_server.serve --model uci_housing_model --thread 10 --p
 | `use_lite` (Only for Intel x86 CPU or ARM CPU) | -    | -       | Run PaddleLite inference                              |
 | `use_xpu`                                      | -    | -       | Run PaddleLite inference with Baidu Kunlun XPU        |
 | `precision`                                    | str  | FP32    | Precision Mode, support FP32, FP16, INT8              |
-| `use_calib`                                    | bool | False   | Only for deployment with TensorRT                     |
+| `use_calib`                                    | bool | False   | Use TRT int8 calibration                              |
 | `gpu_multi_stream`                             | bool | False   | EnableGpuMultiStream to get larger QPS                |
 
 #### 异步模型的说明

diff --git a/python/paddle_serving_server/server.py b/python/paddle_serving_server/server.py
@@ -564,7 +564,7 @@ def run_server(self):
                     "-num_threads {} " \
                     "-port {} " \
                     "-precision {} " \
-                    "-use_calib {} " \
+                    "-use_calib={} " \
                     "-reload_interval_s {} " \
                     "-resource_path {} " \
                     "-resource_file {} " \